Seismic reflection data for ODP Leg 207 and the Demerara Rise come from three principal sources.
A suite of four industry multichannel seismic reflection lines transects the study area. Shell International acquired these data from the survey vessel Petrel (March 1974). Line numbers are C2206, C2207, C2211, and C2212. These data were acquired with a 19.6-L (1200 in3) air gun array of 14 guns towed at a depth of 7–9 m and a Seismic Engineering Multidyne streamer consisting of 60 channels over 3298 m towed at a depth of 15–20 m. The shotpoint interval was 50 m, and data were sampled at 4 ms over a 10-s window length. Original data quality was poor. Reprocessing by the Federal Institute for Geosciences and Natural Resources, Germany, dramatically improved the quality with new semblance analysis, stack, deconvolution, FK-migration, and time-varying bandpass filter. Figure F2A shows a typical frequency spectrum for these processed data.
In excess of 700 line km of seismic reflection data were acquired during the Meteor 49-4 expedition (April–May 2001) on Demerara Rise. All seismic profiling activities during the cruise included simultaneous operation of air gun multichannel seismic reflection and Parasound subbottom sonar profiling. All geophysical data sets were acquired digitally.
High-resolution multichannel seismic reflection data were acquired with the University of Bremen multichannel seismic system. The seismic sound source was a Seismic Systems, Inc. generator-injector (GI) air gun with reduced chamber volume (2 x 0.41 L; 100–500 Hz) fired at a time interval between 9 and 11.5 s and with an air pressure of ~1500 psi. The injector was fired with a 30-ms time delay from the generator shot, which essentially eliminated the bubble pulse. The gun was towed 1.4 m below sea surface. The multichannel seismic streamer (SYNTRON) includes six 100-m-long active sections, each with 16 hydrophone groups. Streamer tow depth was controlled to 3 m below sea level by nine digibirds distributed along its length. Figure F2B shows a typical frequency spectrum for these data.
The Parasound system works both as a high-frequency narrow-beam sounder to determine the water depth and as a low-frequency sediment echo sounder. It makes use of the parametric effect, which produces additional frequencies through nonlinear acoustic interaction of finite amplitude waves. If two sound waves of a certain frequency (e.g., 18 and 22 kHz) are emitted simultaneously, a signal of the difference frequency (e.g., 4 kHz) is generated. The new signal component is traveling within the emission cone of the original high-frequency waves, which are limited to an angle of only 4° for the equipment used. Therefore, the footprint size of 7% of the water depth is much smaller than that for conventional systems, and, as a consequence, both vertical and lateral resolution are significantly improved.
Parasound signals were digitally sampled by the ParaDigMA software at a frequency of 40 kHz with a typical window length of 266 ms. The source signal was a band-limited 2- to 6-kHz sinusoidal wavelet with a 4-kHz dominant frequency having a duration of 2 periods (total length = ~500 µs). Data were filtered with a bandpass filter (3.0–5.0 kHz) and amplitude normalized with a time-varying gain.
One 45-km-long single-channel seismic reflection line was acquired by the JOIDES Resolution while coming to location on the Demerara Rise. The intent was to provide a seismic profile crossing Site 1258 orthogonal to an existing line (line GeoB-221) and allowing improved seismic stratigraphic control by tying to opposing lines (lines GeoB-206 and 215) (Fig. F1). The seismic source used was a Seismic Systems, Inc. GI gun fired in harmonic mode with 2 x 1.72 L chambers. A 56-ms delay firing between the generator and the injector was employed to dampen the bubble pulse. The gun was towed 2 m below the sea surface and fired every 7 s. The receiving element was a Teledyne model 178 single-channel streamer. It consisted of one group of 60 hydrophones with 1.5-m spacing between phones. Towing depth was ~12 m below the sea surface, which, in retrospect, was too deep and caused severe frequency notching in the bandwidth of interest (Fig. F2C). At the time of acquisition, data were bandpass filtered at 20–750 Hz and digitally sampled at a 0.5-ms interval for a 3-s window length.